PolyAnalysis: a deep learning–based framework for nanopore-based poly(A) tail estimation and 3′-end regulation profiling
Abstract
Background: Polyadenylation at the 3′ end of mRNA regulates transcript stability, export, and translation, but poly(A) tail length and polyadenylation-site usage remain difficult to resolve from nanopore sequencing data because of homopolymer-associated errors and complex 3′ end architectures. We developed PolyAnalysis, a signal-level computational framework for estimating poly(A) tail length from raw Oxford Nanopore Technologies signals and linking per-read estimates to confidence-aware analyses of polyadenylation sites, alternative polyadenylation, and repetitive-element-associated 3′ ends.
Results: PolyAnalysis combines CNN–BiLSTM-based region classification with multi-task connectionist temporal classification decoding and adenine-aware training and decoding to infer poly(A)-proximal boundaries, tail length, and model-derived sequence features. Benchmarking against Nanopolish, tailfindr, and Dorado on controlled RNA002 and RNA004 datasets showed competitive tail-length accuracy and high callability. Component-wise ablation and bias-control analyses indicated that region classification, multi-task optimization, adenine-weighted loss, and A-rich decoding jointly contributed to performance. Application to human cell-line and vertebrate transcriptome datasets identified dataset-level variation in tail-length distributions, gene-level heterogeneity, and weak associations with steady-state transcript abundance. Confidence-based filtering separated database-matched, high-confidence novel-candidate and low-confidence polyadenylation sites, while ambiguity-aware endogenous-retrovirus analyses distinguished locus-level from family-level evidence.
Conclusions: PolyAnalysis provides a benchmarked framework for nanopore-based poly(A) tail-length estimation and a confidence-aware basis for exploratory 3′ end profiling. Tail-length estimation is the most extensively validated output, whereas sequence-composition, PAS/APA, and ERV-related results remain dependent on model assumptions, sequencing chemistry, and annotation quality and require further orthogonal validation.
Related
- Active-learning-guided optimization of cell-free systems for genome-wide transcriptomic profiling reveals progressive layers of regulation
- Population-scale single-nucleus multi-omic profiling of skeletal muscle reveals extensive context-specific genetic regulation
- Proteomic Profiling of Endothelial Cells Under Laminar Shear Stress Confirms the Importance of KLF4 in the Regulation of Membrane Protein Expression Compared to Oscillatory Flow
- RiboBA: a bias-aware probabilistic framework for robust ORF identification across diverse ribosome profiling protocols
- Roadmap for inferring metabolic regulation using multiomics, machine learning, and genome-scale modeling.
BibTeX
@article{tian2026polyanalysis,
title = {{PolyAnalysis: a deep learning–based framework for nanopore-based poly(A) tail estimation and 3′-end regulation profiling}},
author = {Qiuxiang Tian and Bosheng Song and Quan Zou and Yansu Wang},
journal = {BMC Biology},
year = {2026},
month = oct,
doi = {10.1186/s12915-026-02749-7},
url = {https://link.springer.com/10.1186/s12915-026-02749-7},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - PolyAnalysis: a deep learning–based framework for nanopore-based poly(A) tail estimation and 3′-end regulation profiling
AU - Qiuxiang Tian
AU - Bosheng Song
AU - Quan Zou
AU - Yansu Wang
PY - 2026
DA - 2026/10/05
JO - BMC Biology
DO - 10.1186/s12915-026-02749-7
UR - https://link.springer.com/10.1186/s12915-026-02749-7
AB - Background: Polyadenylation at the 3′ end of mRNA regulates transcript stability, export, and translation, but poly(A) tail length and polyadenylation-site usage remain difficult to resolve from nanopore sequencing data because of homopolymer-associated errors and complex 3′ end architectures. We developed PolyAnalysis, a signal-level computational framework for estimating poly(A) tail length from raw Oxford Nanopore Technologies signals and linking per-read estimates to confidence-aware analyses of polyadenylation sites, alternative polyadenylation, and repetitive-element-associated 3′ ends. Results: PolyAnalysis combines CNN–BiLSTM-based region classification with multi-task connectionist temporal classification decoding and adenine-aware training and decoding to infer poly(A)-proximal boundaries, tail length, and model-derived sequence features. Benchmarking against Nanopolish, tailfindr, and Dorado on controlled RNA002 and RNA004 datasets showed competitive tail-length accuracy and high callability. Component-wise ablation and bias-control analyses indicated that region classification, multi-task optimization, adenine-weighted loss, and A-rich decoding jointly contributed to performance. Application to human cell-line and vertebrate transcriptome datasets identified dataset-level variation in tail-length distributions, gene-level heterogeneity, and weak associations with steady-state transcript abundance. Confidence-based filtering separated database-matched, high-confidence novel-candidate and low-confidence polyadenylation sites, while ambiguity-aware endogenous-retrovirus analyses distinguished locus-level from family-level evidence. Conclusions: PolyAnalysis provides a benchmarked framework for nanopore-based poly(A) tail-length estimation and a confidence-aware basis for exploratory 3′ end profiling. Tail-length estimation is the most extensively validated output, whereas sequence-composition, PAS/APA, and ERV-related results remain dependent on model assumptions, sequencing chemistry, and annotation quality and require further orthogonal validation.
SN - 1741-7007
ER -